mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Changyuan Lyu <changyuanl@google.com>
To: akpm@linux-foundation.org, linux-kernel@vger.kernel.org
Cc: anthony.yznaga@oracle.com, arnd@arndb.de, ashish.kalra@amd.com,
	 benh@kernel.crashing.org, bp@alien8.de, catalin.marinas@arm.com,
	 corbet@lwn.net, dave.hansen@linux.intel.com,
	devicetree@vger.kernel.org,  dwmw2@infradead.org,
	ebiederm@xmission.com, graf@amazon.com, hpa@zytor.com,
	 jgowans@amazon.com, kexec@lists.infradead.org, krzk@kernel.org,
	 linux-arm-kernel@lists.infradead.org, linux-doc@vger.kernel.org,
	 linux-mm@kvack.org, luto@kernel.org, mark.rutland@arm.com,
	mingo@redhat.com,  pasha.tatashin@soleen.com,
	pbonzini@redhat.com, peterz@infradead.org,  ptyadav@amazon.de,
	robh@kernel.org, rostedt@goodmis.org, rppt@kernel.org,
	 saravanak@google.com, skinsburskii@linux.microsoft.com,
	tglx@linutronix.de,  thomas.lendacky@amd.com, will@kernel.org,
	x86@kernel.org,  Changyuan Lyu <changyuanl@google.com>
Subject: [PATCH v8 13/17] x86/boot: make sure KASLR does not step over KHO preserved memory
Date: Fri,  9 May 2025 00:46:31 -0700	[thread overview]
Message-ID: <20250509074635.3187114-14-changyuanl@google.com> (raw)
In-Reply-To: <20250509074635.3187114-1-changyuanl@google.com>

From: Alexander Graf <graf@amazon.com>

During kexec handover (KHO) memory contains data that should be
preserved and this data would be consumed by kexec'ed kernel.

To make sure that the preserved memory is not overwritten, KHO uses
"scratch regions" to bootstrap kexec'ed kernel. These regions are
guaranteed to not have any memory that KHO would preserve and are used as
the only memory the kernel sees during the early boot.

The scratch regions are passed in the setup_data by the first kernel with
other KHO parameters. If the setup_data contains the KHO parameters, limit
randomization to scratch areas only to make sure preserved memory won't get
overwritten.

Since all the pointers in setup_data are represented by u64, they require
double casting (first to unsigned long and then to the actual pointer type)
to compile on 32-bits. This looks goofy out of context, but it is
unfortunately the way that this is handled across the tree. There are at
least a dozen instances of casting like this.

Signed-off-by: Alexander Graf <graf@amazon.com>
Co-developed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Co-developed-by: Changyuan Lyu <changyuanl@google.com>
Signed-off-by: Changyuan Lyu <changyuanl@google.com>
---
 arch/x86/boot/compressed/kaslr.c | 50 +++++++++++++++++++++++++++++++-
 1 file changed, 49 insertions(+), 1 deletion(-)

diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c
index f03d59ea6e40f..3b0948ad449f9 100644
--- a/arch/x86/boot/compressed/kaslr.c
+++ b/arch/x86/boot/compressed/kaslr.c
@@ -760,6 +760,49 @@ static void process_e820_entries(unsigned long minimum,
 	}
 }
 
+/*
+ * If KHO is active, only process its scratch areas to ensure we are not
+ * stepping onto preserved memory.
+ */
+static bool process_kho_entries(unsigned long minimum, unsigned long image_size)
+{
+	struct kho_scratch *kho_scratch;
+	struct setup_data *ptr;
+	struct kho_data *kho;
+	int i, nr_areas = 0;
+
+	if (!IS_ENABLED(CONFIG_KEXEC_HANDOVER))
+		return false;
+
+	ptr = (struct setup_data *)(unsigned long)boot_params_ptr->hdr.setup_data;
+	while (ptr) {
+		if (ptr->type == SETUP_KEXEC_KHO) {
+			kho = (struct kho_data *)(unsigned long)ptr->data;
+			kho_scratch = (void *)(unsigned long)kho->scratch_addr;
+			nr_areas = kho->scratch_size / sizeof(*kho_scratch);
+			break;
+		}
+
+		ptr = (struct setup_data *)(unsigned long)ptr->next;
+	}
+
+	if (!nr_areas)
+		return false;
+
+	for (i = 0; i < nr_areas; i++) {
+		struct kho_scratch *area = &kho_scratch[i];
+		struct mem_vector region = {
+			.start = area->addr,
+			.size = area->size,
+		};
+
+		if (process_mem_region(&region, minimum, image_size))
+			break;
+	}
+
+	return true;
+}
+
 static unsigned long find_random_phys_addr(unsigned long minimum,
 					   unsigned long image_size)
 {
@@ -775,7 +818,12 @@ static unsigned long find_random_phys_addr(unsigned long minimum,
 		return 0;
 	}
 
-	if (!process_efi_entries(minimum, image_size))
+	/*
+	 * During kexec handover only process KHO scratch areas that are known
+	 * not to contain any data that must be preserved.
+	 */
+	if (!process_kho_entries(minimum, image_size) &&
+	    !process_efi_entries(minimum, image_size))
 		process_e820_entries(minimum, image_size);
 
 	phys_addr = slots_fetch_random();
-- 
2.49.0.1015.ga840276032-goog


  parent reply	other threads:[~2025-05-09  7:47 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-09  7:46 [PATCH v8 00/17] kexec: introduce Kexec HandOver (KHO) Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 01/17] memblock: add MEMBLOCK_RSRV_KERN flag Changyuan Lyu
2025-10-10  9:33   ` Breno Leitao
2025-10-13 14:59     ` Pratyush Yadav
2025-10-13 16:40       ` Pratyush Yadav
2025-10-14  8:34         ` Breno Leitao
2025-10-14 13:10           ` Pratyush Yadav
2025-11-05 10:18             ` Breno Leitao
2025-11-06  8:24               ` Mike Rapoport
2025-05-09  7:46 ` [PATCH v8 02/17] memblock: Add support for scratch memory Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 03/17] memblock: introduce memmap_init_kho_scratch() Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 04/17] kexec: add Kexec HandOver (KHO) generation helpers Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 05/17] kexec: add KHO parsing support Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 06/17] kexec: enable KHO support for memory preservation Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 07/17] kexec: add KHO support to kexec file loads Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 08/17] kexec: add config option for KHO Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 09/17] arm64: add KHO support Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 10/17] x86/setup: use memblock_reserve_kern for memory used by kernel Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 11/17] x86/kexec: add support for passing kexec handover (KHO) data Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 12/17] x86/e820: temporarily enable KHO scratch for memory below 1M Changyuan Lyu
2025-11-24 19:24   ` Usama Arif
2025-11-25  0:56     ` H. Peter Anvin
2025-11-25 12:23       ` Pratyush Yadav
2025-11-25 13:53       ` Mike Rapoport
2025-11-25 13:15     ` Pratyush Yadav
2025-11-25 13:50       ` Mike Rapoport
2025-11-25 18:47         ` Usama Arif
2025-11-26  6:14           ` Mike Rapoport
2025-11-26  7:25             ` Usama Arif
2025-11-25 14:31       ` Usama Arif
2025-11-25 14:39         ` Pratyush Yadav
2025-05-09  7:46 ` Changyuan Lyu [this message]
2025-05-09  7:46 ` [PATCH v8 14/17] x86/Kconfig: enable kexec handover for 64 bits Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 15/17] memblock: add KHO support for reserve_mem Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 16/17] Documentation: add documentation for KHO Changyuan Lyu
2025-05-09  7:46 ` [PATCH v8 17/17] Documentation: KHO: Add memblock bindings Changyuan Lyu
2025-12-11  3:00   ` Ricardo Neri

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250509074635.3187114-14-changyuanl@google.com \
    --to=changyuanl@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=anthony.yznaga@oracle.com \
    --cc=arnd@arndb.de \
    --cc=ashish.kalra@amd.com \
    --cc=benh@kernel.crashing.org \
    --cc=bp@alien8.de \
    --cc=catalin.marinas@arm.com \
    --cc=corbet@lwn.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=devicetree@vger.kernel.org \
    --cc=dwmw2@infradead.org \
    --cc=ebiederm@xmission.com \
    --cc=graf@amazon.com \
    --cc=hpa@zytor.com \
    --cc=jgowans@amazon.com \
    --cc=kexec@lists.infradead.org \
    --cc=krzk@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=ptyadav@amazon.de \
    --cc=robh@kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=rppt@kernel.org \
    --cc=saravanak@google.com \
    --cc=skinsburskii@linux.microsoft.com \
    --cc=tglx@linutronix.de \
    --cc=thomas.lendacky@amd.com \
    --cc=will@kernel.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®